People love to game. This was clear in the Before Times, and the past year has made it glaringly obvious. The lockdowns and all the time people are spending at home has accelerated an already strongly growing adoption curve. This has become one of those truths that are so obvious we tend to ignore it, the way fish do not think about water. But this belies the scale of gaming overall and the increasing load it puts on all the systems required to make it work.
Take two data points as example. According to Limelight Networks, a content delivery network provider, the average gamer spent close to seven hours a week gaming last year. According to consultancy DFC Intelligence (who admittedly we know nothing about) says 3 billion people played video games last year. By contrast, network equipment maker Cisco only briefly mentions gaming in its annual research report on networking trends, lumping it in with social media, business applications and IoT.
Much of the networking industry focuses on video traffic which is admittedly a huge amount of Internet traffic. However, there is good reason to believe that everyone is undercounting gaming traffic. Bandwidth management company Whistleout regularly publishes stats on gaming data usage. By their count, the average gaming app consumes between 3MB and 300 MB of data per hour. In the chart below, each of those titles looks tiny when compared to Netflix users’ data usage. However, Netflix has 183 million subscribers, gaming has 3 billion users. Working through the math this comes to 455 Petabytes a day for Netflix and 450 PT a day for gaming. The gaming figure only includes people playing games, it does not include people watching games on other platforms like Twitch and YouTube.
In Cisco’s defense, it is hard to distinguish what is gaming traffic in their studies. Gaming traffic is fragmented across thousands of titles on PCs, consoles and phones. It is highly likely that the figures we have for gaming hugely understate actual data traffic driven by gaming. And this is still early days. The number of gamers continues to grow as do the levels of engagement and time spent on gaming. If you factor in emergent game formats like VR and AR, which are incredibly data intensive, it is easy to see an Internet in which gaming starts to rival video as a leader in data usage.
When viewed in this light, we think the industry needs to understand what this means for the networks underpinning all that data. Gaming is a major driver of Internet traffic but has its own usage models that require understanding as separate from other forms of data which garner far more attention. In coming weeks we are going to take a deep look into these challenges and the opportunities they present.
Photo by JESHOOTS.COM on Unsplash
Thanks – interesting argument!
A couple more drivers worth considering:
1. What % of games are played online vs offline?
2. How much bandwidth is consumed by digital distribution of games – ie downloading newly purchased games and content?
Good points. I don’t have an answer at hand for question 1. I imagine it is fairly high, but one of things I will explore in a future post is the fact that game companies really want people to play online, it drives engagement and conversion. So what happens when players have bad Internet connections and blame the game company? I think that is a major source of user frustration.
As for #2, that is a fair size element, is fairly small when compared to a streaming game. For a lot of games, the file size itself is probably equivalent to a few sessions of online play
Pingback: Choose Your Fighter | Digits to Dollars·